1. sports
2. fishing
3. freshwater fishing

BIOLOGY
OF
REPRODUCTION
39,
1039-1050
F Prostaglandins
(1988)
Function
as Potent
the Postovulatory
PETER
W. SORENSEN,2’3
Female
J.
TOSHIAKI
HARA,’
Olfactory
Stimulants
Sex Pheromone
NORMAN
Department
That
Comprise
in Goldfish’
E. STACEY,3
and
FREDERICK
WM.
GOETZ5
of Zoology3
University
of Alberta
Edmonton,
Alberta
T6G 2E9 Canada
Department
of Fisheries
and
Oceans4
Fresh water Institute
Winnipeg,
Manitoba
R3T2N6
Canada
and
of Zoology5
Department
University
of Notre
Notre
Dame,
Indiana
Dame
46556
ABSTRACT
This
they
study
establishes
stimulate
strated
that
Next,
using
male
that
ovulated
spawning
ovulated
and
female
behavior
goldfish
and
recording
F type
the goldfish
female
goldfish
prostaglandin-injected
electro-olfactogram
release
comprise
(EOG),
we
prostaglandins
determined
that
waterborne
potent
olfactory stimulants for mature male goldfish. Prostaglandin
F2a (PGF2)
prostaglandin
F2a (15KPGF2a)
were the most potent prostaglandins;
the former
10
PGF
‘#{176}M
and the latter a detection threshold
metabolites
are an important
component
low concentrations
of waterborne
the odor
of ovulated
fish.
Together
as a preovulatory
“priming”pheromone
may
commonly
serve
of 10’2M.
Studies
of the pheromone.
PGFs
with
as reproductive
spawning
1982;
goldfish,
that
behavior
Stacey
et
like many
teleost
rapidly
stimulates
(Partridge
al.,
1986).
et
The
al.,
prostaglandins
and
function
had
a detection
15KPGF2a
that male
as
15-ketothreshold
indicated
of
that
using the EOG
that are indepengoldfish
exposed
to
elicited
by
hormone
and their
leasing”
actions
of this pheromone
physiological
or “priming”
actions
mone
that
is released
by female
ovulation.
We recently
discovered
fish,
male
1976;
behavioral
where
exposure
functions
metabolites
to
in fish.
INTRODUCTION
Recently
ovulated
release
a pheromone(s)
water
and its metabolite
exhibit
reproductive
behaviors
similar
to those
our recent
discovery
that a steroidal
maturational
for goldfish,
these findings
suggest
that hormones
pheromones
to the
of prostaglandin-injected
fish
Cross-adaptation
experiments
have separate olfactory receptor
sites for PGF2a
other
olfactory
stimulants.
Finally,
we established
demonstrated
that goldfish
dent
from
those
that detect
(PGFs)
postovulatory
pheromone.
We first demo nrelease immunoreactive
PGFs to the water.
Liley,
or
goldfish
mone,
“re-
release
the oocyte
maturational
1 7a ,2013-dihydroxy-4-pregnen-3
to the
water
where
it functions
Waterborne
17,20P
stimulates
the
Accepted
July
13, 1988.
Received
March
1, 1988.
‘This
study
was supported
by the Alberta
Heritage
Foundation
for
Medical
Research
(Fellowship
to P. W. S.), the Department
of Fisheries
and
Oceans
Canada
(Contract#
FP430-6-9064/01-1SF
to P. W. S. and
T. J. H.),
the
Natural
Sciences
and
Engineering
Council
of Canada
(Grant#
A7576
to T. J. H. and #A2903
to N. E. S.), and the National
Science
Foundation
(Grant#DcB-85
17718
to F. W. G.).
2 Reprint
requests
and present
address:
Peter
W. Sorensen,
Department
of Fisheries
and Wildlife,
University
of Minnesota,
200
Hodson
Hall,
1980
FoIwell
Avenue,
St. Paul, MN 55108.
of
mature
males
via
increase
milt
(sperm
tion
sen,
by the time
1986;
Dulka
1987).
In this
fish postovulatory
signal
1039
contrast
with
the
of another
pherogoldfish
prior
to
that
preovulatory
that
study,
differentiates
-one
steroid
hor(17 ,20P),
as a pheromone.
endocrine
system
their
olfactory
system
to
and
seminal
fluid)
produc-
of spawning
(Stacey
and
et al.,
1987;
Sorensen
we hypothesized
pheromone
(the
ovulated
that
specific
from
Sorenet al.,
the goldchemical
nonovulated
1040
SORENSEN
goldfish)
is likely
compound
17,20P
is
to
be
a hormone
or hormone-like
closely
associated
with
a
candidate
for
poor
ovulation
the
pheromone
because
its synthesis
declining
by the
time
of ovulation
1987),
and
because
waterborne
minor
effects
on male
behavior
itself.
postovulatory
and
release
are
(Dulka
et a!.,
17,20P
has only
(Sorensen,
unpublished
Kittredge
et a!. (1971)
first suggested
that aquatic
organisms
are likely
to have evolved
to use hormones
and
their
metabolites
as sex pheromones
because
they represent
pre-existing
and relevant
chemical
cues
whose
detection
need
only involve
an externalization
of
internal
hormone
receptor
mechanisms.
More
recently,
Doving
(1976)
suggested
that this possibility
also apply
have since
for teleost
implicated
AL.
tory pheromone
because,
where
studied
in mammals,
PGs have been
found
to be rapidly
metabolized
and
excreted
(Samuelsson
et al., 1975;
Hoult
and Moore,
1977;
Granstrom
is known
about
duration
of the
and Kindahl,
1982).
Although
little
PG metabolism
in fish, both the short
spawning
response
elicited
by PGF2a
injection
(Stacey
and Goetz,
indomethacin
(a cyclo-oxygenase
results).
might
studies
ET
fish.
Although
several
steroidal
pheromones
in
quickly
blocks
1976)
rapidly.
indicate
Several
spawning
that
1982)
in ovulated
goldfish
predictions
and
inhibitor)
also
arose
from
the
fact that
treatment
females
(Stacey,
metabolize
our
PGF2a
hypothesis
that
prostaglandins
and/or
their
metabolites
function
as
the goldfish
postovulatory
pheromone.
First,
recently
ovulated
and PGF2 a-injected
goldfish
should
release
PGFs
to the water.
Second,
the olfactory
sense
of
mature
males
should
detect
waterborne
prostaglan-
fish (Colombo
et a!., 1982,
van den Hurk
and Lambert,
1983;
Lambert
et al., 1986;
van den Hurk
et
a!., 1987;
Stacey
et al., 1987),
no study,
with
the
exception
of
ours
on
the
goldfish
preovulatory
pheromone,
has simultaneously
demonstrated
pheromone
release,
olfactory
sensitivity,
and a relevant
biological
response
(i.e.,
a neuroendocrine
reflex
or
dins
because
olfactory
ablation
blocks
the responsiveness
of goldfish
to pheromones
(Partridge
et al.,
1976;
Stacey
and Kyle,
1983).
Third,
PGF2a-injected
goldfish
should
release
a potent
olfactory
stimulant(s)
sexual
arousal).
identified
a fish
We suspected
should
be relatively
specific
and not respond
to other
odorants.
Fifth,
mature
male
goldfish
exposed
to
physiological
concentrations
of waterborne
PGFs
Similarly,
postovulatory
for several
no
previous
pheromone.
reasons
that
postovulatory
pheromone
could
glandins
(PGs).
First,
circulating
F type
prostaglandins
(PGFs)
(and
other
te!eosts)
fard,
1979:
Cetta
probably
reflecting
rupture
Second,
apparently
on the
ovulated
and
a
the
time
the
goldfish
consist
of prostaand ovarian
levels of
increase
in goldfish
of ovulation
Goetz,
role
in
has
(Bouf-
1982;
Goetz,
1983),
modulating
follicular
detected
by olfactory
receptors
that
also
to prostaglandins.
Fourth,
those
olfactory
that
respond
to waterborne
prostaglandins
should
exhibit
the
same
reproductive
elicited
by exposure
to water
from
ovulated
This
study
sought
to test these
predictions
variety
water
of techniques,
including
samples,
electrophysiological
behavioral
1979;
Stacey
as a hormonal
behavior
and
Goetz,
signal
that
in goldfish
brain
(Stacey
and
goldfish
injected
Peter,
with
exhibit
normal
female
spawning
1976;
Stacey
and Goetz,
1982),
odor that elicits
male reproductive
MATERIALS
1982;
triggers
Goetz,
female
Prostaglandin
direct
actions
and
through
1979).
PGF2a
behavior
but also
behaviors
Third,
not
nononly
(Stacey,
release
an
identical
to those
elicited
by the odor
of ovulated
females
(Sorensen
et al.,
1986).
Spawning
males
do not
distinguish
between
naturally
ovulated
females
and
PGF2ainjected
females
(Stacey,
1981).
In particular,
we believed
that PGF2a
metabolites
might
be an important
component
of the postovula-
Release
Prostaglandin-Injected
We
female
behaviors
females.
using
radioimmunoassay
recording,
a
of
and
observation.
(Dennefors
et
a!.,
1983;
Goetz,
1983).
circulating
prostaglandin
F2 a (PGF2 a),
the
most
abundant
PGF
in goldfish
(Bouffard,
1983),
acts
spawning
at
study
that
is
respond
receptors
first
sought
to
goldfish
release
AND
by
METHODS
Ovulated
Goldfish
determine
PGFs
to
whether
the water.
ovulated
Mature
vitellogenic
females
(Ozark
Fisheries,
Stoutland,
MO)
were
induced
to ovulate
using an established
protocol
(Stacey
et a!., 1979).
Briefly,
they were moved
from
stock
tanks
(14#{176}C,16L:8D
photoperiod,
lights
on at
0800
h)
into
70-1
flow-through
aquaria
(20#{176}C;
16L:8D)
containing
aquatic
vegetation
(spawning
substrate)
at 2100
h on Day
1; on the morning
of
Day
3, they
were checked
for ovulation
by applying
gentle
pressure
to their abdomens.
Ovulated
fish were
PROSTAGLANDIN
divided
into
from
males
to spawn
2 groups:
(“ovulated”)
with
released
ovulated,
one group
and the
males
for
several
was
other
hours
all their
eggs (“spawned-out”).
and spawned-out
fish were
individual
glass
aerated
jars
water.
containing
After
until
they
was
released/excreted
female
i.m. with either
10 pg
(an amount
equivalent
had
sensitivity
odorants
nares
and
1041
because
olfactory
tranquilizing
epithelia
by
using
an
collected
by
goldfish
entially
bridged
diameter
positioned
female
pheromone
the
release
1986]),
[Stacey
and
Goetz,
1982;
Sorensen
et al.,
or buffer
alone,
placed
into jars,
and their
doses
of anesthetic
reduce
olfactory
Five-second
pulses
the flow perfusing
apparatus
designed
to
fluctuations
response
was
reference
electrode
was
procedures).
(Sorensen
positioned
minimize
(Evans
and
recorded
differ-
et al., 1987),
at the location
skin
recording
electrode
was
that
produced
a maximal
3.5 with
0.1 N HCI,
which
was extracted
ethyl
acetate
was
and
at 37#{176}C
and
PGF”)
using
Goetz,
1982).
assayed
for
an established
Siliconized
PGF2a
(“immunoprotocol
(Cetta
glassware
was
used
sample
collection,
extraction,
and radioimmunoassay.
Extraction
efficiency
for PGF2a
was estimated
greater
reacts
90%.
than
100%
with
with
PGE1,
The
PGF2a;
PGE2,
1 5-keto-prostaglandin
Olfactory
fish
Sensitivity
to Waterborne
and
PGA2;
F2 a
a metabolite
Ozark
glandins
was
gram
(EOG)
and
of Male
Gold-
for
to be
a concentration
able in
responses
cross0.01%
0.2 3%
of PGF2
of
from
Fisheries)
determined
responses.
mature
which
to
male
sperm
waterborne
by recording
The
EOG
with
A protocol
previously
used
a
goldfish
could
be
prosta-
goldfish
3 mg/kg
Montreal
stand,
and
dechlorinated
their
body
Que.,
gills and
11#{176}Cwater.
wieght;
Canada),
Rhone-Paulene
placed
on
nares
were
perfused
with
These
animals
were
not
series.
106M
to
a
grounded.
Because
PG solutions
methanol
the standard
because
goldfish
is representative
because
we used
it
the
response
L-serine
after
water
control
olfactory
responses
was employed
(Sorensen
et al.,
1987).
Briefly,
mature
male
goldfish
were
immobi!ized
with
an i.m.
injection
of Flaxedil
(gallamine
triethiodide;
Pharmacie,
were
a
Electrical
(Grass
prostaMI) in
and placing
17 !1l
to create
a 106 M
diluted
to create
methanol
(7 X
control
and
odorant
was
104M
were
detect-
methanol),
tested
in
its
as
potency
of other
a standard
as an odorant
for
L-amino
acids and
in earlier
studies
(Sorensen
et al., 1987).
Amino
acids
are important
components
of food
odors
and are potent
olfactory
stimulants
for fish (Caprio,
1984;
Hara,
1986).
Each
test
stimulus
was
tested
3 times,
and
2 mm were
allowed
between
stimuli
for recovery.
Responses
were
averaged
and
then
expressed
as a percentage
of
electro-olfactois a multiunit
to record
fish
standard
water
(no
conjunction
with
each
series
of solutions.
A blank
water
control
(no odorant
added)
and an amino
acid
standard
(10
M L-serine;
Sigma
Chemical
Co., St.
Louis,
MO) were
also tested
at the beginning
and end
of each
concentration
series.
L-Serine
was chosen
as
transepithelial
voltage
transient
recorded
from
the
surface
of
the
olfactory
epithelium
and
reflects
olfactory
receptor
potentials
(Ottoson,
1971; Getchell,
1974).
The
M L-serine
a “blank”
glass-distilled
methanol
at 1 mg/mI
into 50 ml of dechlorinated
water
dilution
that was then
progressively
Prostaglandins
The olfactory
sensitivity
(fish
with
tubercles
and
expressed;
antiserum
employed
24% with
PGF1a;
control.
surface
experiments
signals
were
amplified
by a DC-preamplifier
7Pl) and displayed
on a pen recorder.
Test
solutions
were prepared
by dissolving
glandins
(Cayman
Chemical
Co., Ann
Arbor,
nitrogen
reactive
10
to
the
on the
in previous
response
minimal
and split into 2 aliquots,
each of
with
5 ml of ethyl
acetate.
The
evaporated
under
a stream
of
to the
response
placed
As
water
was
collected
and
frozen
2 h later.
Water
samples
from
both the ovulation
and PGF2 a-injection
experiments
were
later
thawed,
acidified
to a pH of
added)
of
the
using
Ag/AgCl
electrodes
(WPI Type
EH-lF)
to
saline/gelatin-filled
glass
pipettes
(tip
60-80
pm).
The
recording
electrode
was
just above
the olfactory
epithelium,
and
(noninvasive
evoke
and
1985).
into
(Lewis
et a!.,
were
introduced
pressure
and
temperature
Hara,
1985).
The EOG
PGF2a
in 10 p1
to that used to
behavior
IN GOLDFISH
anesthetized
destroy
fish
Nonovulated,
then placed
into
2 h, 10 ml of water
spawning
isolated
allowed
1.5 1 of dechlorinated,
from each jar and frozen.
To
determine
if PGF2a
is
PGFz-injected
fish,
nonovulated
were
injected
of saline
buffer
kept
was
PHEROMONE
10-6 M
portion
a
Responses
PGF2a,
(Stacey
elicited
by the
most
recent
iO
M
responses
to the most
recent
blank
had
been
subtracted.
Responses
to
solutions
of the
were
were
response
tested
PGF3a,
and Goetz,
corrected
attributable
to
remove
that
to methanol.
to all 5 prostaglandins
PGE1,
1982).
and
PGE2)
(PGF1a,
found
in
fish
SORENSEN
1042
Pheromone
and
Release
Olfactory
Sensitivity
To test the
PGF metabolites
cy
of
by PGF2c,.-Injected
water
Fish
to PGF2a
possibility
that
we determined
collected
from
goldfish
olfactory
release
poten-
PGF2a-injected
fish
compared
its potency
to that of PGF2a
added
ly to
the
water.
Twenty-five-gram
females
nonvitellogenic
ovaries
were injected
i.m. with
10 pg PGF2a
immunoreactive
rinsed,
and
(the
PGF
placed
dechlorinated
into
water
sponses
of mature
recorded
using
dose
release)
used
earlier
to
or 10 p1 buffered
beakers
males
the
45
to these
water
established
undiluted
water
evoked
both
water
samples
were
mm.
1.5
1 of
EOG
re-
samples
protocol.
extremely
diluted
and
directwith
either
measure
saline,
containing
(11#{176}C) for
were
Because
large
10 and
AL.
tory
also
Metabolites
female
the
ET
receptors
respond
to the
stimulus.
adapting
stimulus
We adapted
the
olfactory
epithelia
of male goldfish
PGF2a
or 107M
15K-PGF2a,
and
tions
of salineand PGF2a-injected
to either
10 8 M
tested
10% dilufemale
water
to
see if adaptation
to PGFs
would
selectively
responses
elicited
by
PGF2a-injected
fish
Responses
were
analyzed
in mV rather
than
reduce
water.
relative
to L-serine,
because
it could
not
be assumed
that
responses
to the L-serine
standard
were
not
influenced
by adaptation.
Adaptation
was initially
verified
by demonstrating
that
and 108M
PGF2a
equal
concentrations
Although
relationships
responses,
100 times
responding
the test
to
responses
to 1 0
were
abolished
of themselves.
the different
to PGF2a
M 1 5K-PGF2a
by
adaptation
to
EOG concentration-response
and its metabolites
suggested
with
dechlorinated
water
for testing.
Responses
to
salineand PGF2 a-injected
fish water
were expressed
relative
to a 10
M L-serine
control
and were
compared
using paired
t-tests.
Because
the
EOG
responses
elicited
by PGF2a-
that
goldfish
have
more
than
one class of olfactory
receptors
for PGFs,
the specificity
of these
receptors
remained
to be demonstrated.
Therefore,
the cross-
injected
fish water
were
too large to be attributable
to PGF2a
alone
(see
Results),
we next
determined
EOG
responses
of mature
males
to PGF2a
metabo-
established
categories
of potent
in fish (Hara,
1986;
Sorensen
sentative
amino
acid
(105M
lites.
fish
(10
The only known
study
of PGF2a
metabolism
in
found
that
goldfish
ovaries
produce
15-keto-
prostaglandin
results).
F2a
in
vitro
1 5-Keto-prostaglandin
and
13,
14-dihydro1445KPGF2a)
are
PGF2a
in a variety
1975;
Kindahl,
Hoult
1982).
determined
arachidonic
Specificity
Receptors
PGFs, and
(Goetz,
F2a
unpublished
(1 5KPGF2
a)
1 5-keto-prostaglandin
F2 a (13,
also
the
initial
metabolites
of
of mammals
(Samuelsson
et a!.,
and
Moore,
EOG responses
to both
of these
acid, the precursor
of Prostaglandin
to PGF2 a-Injected
Other
Odorants
1977;
Granstrom
of mature
males
and
were
compounds
of PGF2a.
as to
as well
Olfactory
Fish
Water,
To verify
whether
the EOG
response
to PGF2ainjected
fish water
was attributable
to the olfactory
receptors
responding
to PGFs
a cross-adaptation
experiment
was conducted.
In cross-adaptation,
the
EOG response
to a test odorant
is first measured,
and
then
the
olfactory
epithelium
is perfused
with
an
adapting
odorant
to which
pulses
of the test odorant
(made
up in the adapting
stimulus)
are added
(see
Caprio
and Byrd,
1984).
Reductions
in EOG response
magnitude
to the test odorant
during
adaptation
are
interpreted
as reflecting
the extent
to which
olfac-
adaptation
experiments
variety
of PGs
were
as well
M taurocholic
Sigma
Chemical
Sigma
Chemical
similarity
between
carboyxl
group,
acid-which
is a
acid
we
to
made
up
as a sodium
salt;
for different
(Hara,
1982;
did not test
specifically
amino
Caprio
acids are
and Byrd
other
amino
acids.
whether
goldfish
Last,
to
possess
for PGF2a
and
of each
comadapted
to the
PGF.
Behavioral
fish
a
three
olfactory
stimulants
et a!., 1987):
a repreL-serine),
bile
acid
separate
classes
of olfactory
receptors
15KPGF2a,
various
concentrations
pound
were
tested
while
a fish was
other
include
of the
Co.),
and a steroid
(108M
17,20P;
Co.).
Because
the only
structural
PGs and amino
acids
is a terminal
a feature
also shared
by arachidonic
poor
olfactory
stimulant,
and the
receptor
types
proposed
not mutually
exclusive
1984),
determine
extended
as representatives
Responses
to Waterborne
Finally,
determine
chasing
ovipores
mone
females
weeds
a behavioral
when
experiment
male
exposed
behavior
of females
and sides
source;
allow
where,
Gold-
Pros taglandins
whether
behavior
reproductive
of Male
(as
goldfish
was
exhibit
to waterborne
is characterized
and
constant
if trying
to
conducted
to
reproductive
PGFs.
Male
by extensive
nudging
identify
of
the
their
phero-
Partridge
et a!., 1976).
If receptive,
themselves
to be pushed
into
aquatic
in a reflexive
action,
they
oviposit
in
PROSTAGLANDIN
synchrony
male(s)
1976;
with
the
of sperm
groups
and/or
of males
nonovulated
1976;
personal
the
odor
of
females
become
the
odor
occasionally
females
and
interact
pursuing
(Partridge
other
et a!.,
with
visual
images
of
females
(Partridge
et a!., 1976;
Sorensen
et a!., 1986).
Grouped
males
exposed
to
these
odors
exhibit
dramatic
increases
in nudging
and
chasing;
these
behaviors
are not elicited
by exposure
to food
odor,
the
odor
of nonovulated
females,
or the odor
of
waterborne
17, 20P (Sorensen
et a!., 1986).
The
here
protocol
of the behavioral
is based
directly
on that used
experiments
in earlier
grouped
males
(Sorensen
et al., 1986).
not included
in these
groups
because
tive
females
would
act
and
trolled
elicits
behavioral
variable
that
some
chasing.
Tests
offered
nonreceptive
a controlled,
could
test
by exposure
of
5 males
gravel
number
a 1 5-mm
presumably
of nudges
pre-test
(10-6 M or less) was
the aquaria
through
experimental
period
an uncon-
held
with
which
in
flow-
for
(social
contact)
were
during
which
an ethanol
pumped
aerators
started
of
midline
on the tank
of times
fish picked
searching
food),
Release
Approximately
PGF
was
found
20 times
more
in water
collected
and
observed
control
by peristaltic
pump
into
at lOml/min.
A 15-mm
when
the input
solution
reactive
those
of the
PGF were also found
in water
samples
fish injected
with
PGF2a;
approximately
injected
PGF2a
had been released.
Olfactory
Sensitivity
Goldfish
to Prostaglandins
The
olfactory
epithelium
from
PGF2a
at
bond;
group
concentrations
longer
observed.
ranks
test
for
and
were
repeated
progressively
using
difdecreasing
of PGFs
until
responses
were
no
Results
were analyzed
by the signeddependent
samples.
goldfish
was
one
position
be saturating,
those
elicited
response
differ
(PGF1a
curves
structurally
has
an
for
extra
PGF
lacks a double
bond;
PGE2
substituted
for a hydroxyl
group),
has a
were
similar
to the PGF2a
curve in shape
but shifted
to the
right
by one
log unit.
PGE1,
which
differs
from
PGF2a
at two
positions,
was the least
stimulatory
(Fig.
la).
A structural
isomer
of PGF2a,
11I3PGF2a,
was
also
considerably
and its response
(data
not shown).
sure
and
less
to PGs had relatively
small tonic
components,
similar
odorants
to
the
(Fig.
5po.o1
than
EOGs
large
but
elicited
by
to that
elicited
exposure
F prostaglandin
period
(x ± S EM).
(PGF)
to
compared
released
PGF
(ng)
(10)
5)
1.6
23.8±
2.1
2.3
142.8
(
compared
(ng)
Sample
size
( 6)
( 2)
when
of
by
PGE1
expo-
phasic
components
were fundamentally
(
when
PGF2a,
other
2).
Nonovulators
Ovulated
Spawned-out
Saline-injected
PGF20-injected
+p,005
stimulatory
curve
was similar
EOG
responses
Group
procedures
of fish
male
appeared
to
than
3 times
M L-serine.
Concentration
PGF3a,
and PGE2,
which
tration
was
injected,
dye
ly encountered
testing
groups
of
acutely
sensitive
to all PGs tested,
especially
PGF2a
(Fig. la). PGF2a
had a detection
threshold
of approximately
10’#{176}M, and at a concentration
of 106M,
TABLE
1. Immunoreactive
male goldfish
during
a 2-h
These
ferent
from
2%
of Male
was switched
to 108M
PGF2a,
108M
15KPGF2,
or another
ethanol
control.
Observers
were unaware
of the treatment
given.
Although
final tank
concenonly
0.4%
that
of the
concentration
injection
indicated
that males repeated“wisps”
of slightly
diluted
odorant.
immunoreactive
from
ovulated
females
than
in water
from either
nonovulated
fish or
spawned-out
fish who released
equivalent,
marginally
detectable
levels
(Table
1). High
levels
of immuno-
double
ketone
responses
elicited
were
reproductive
overnight
Pros taglandin
by 10
PGF1a,
we
(20#{176}C;16L:8D)
that contained
vegetation
(spawning
substrate).
swimming
activity
(number
fish
crossed
a vertical
feeding
activity
(number
up bottom
the
for
means
were
1043
where
response
magnitude
it elicited
responses
more
pheromone
represent
repeatable
through
70-I aquaria
gravel
and aquatic
The
next
morning,
were
recep-
we know
occasionally
using
grouped
males
whether
the behavioral
to waterborne
PGFs
behaviors.
Groups
times
glass),
females
used
tests of
Females
sexually
as an independent
source,
GOLDFISH
RESULTS
observations).
Single
males exposed
to
ovulated
females
or PGF2a-injected
more
active,
feed less, are attracted
source,
IN
by a companion
(Breder
and
Rosen,
1966;
Partridge
et a!.,
Stacey
and Kyle,
1983).
Spawning
is chaotic
with
males
to
release
PHEROMONE
3)
to nonovulators
to saline-injected
and
fish.
spawned-out
±
±
±
±
by
0.19
7.50
0.51
0.48
2i.74
fish.
fe-
1044
SORENSEN
300
(c)
400
B
[I
C)
AL.
(b)
400
400
ET
Saline-injected
fish water
300
#{149}13,14-15K-PGF2z
300
(I)
PGF2.injected
C
0
PGF2
#{149}
15K-PGF2ca
#{149}
Arachidonic
fish water
Acid
**
0.
Cl)
200
C)
200
200
C)
>
100
CC
100
**
C)
0
100
B
AB
0
0
-11
-9
Log
Molar
-7
-5
0.01
-13
0.1
Concentration
Concentration
-11
Log
-9
Molar
-7
-5
Concentration
FIG.
1. (a) Semi-logarithmic
plot
of the electro-olfactogram
recording
(EOG)
responses
of male
goldfish
to waterborne
prostaglandins
(PGs).
Average
response
magnitude
is represented
as a percentage
of that induced
by io
M L-serine.
Vertical
bars represent
standard
error.
Abbreviations
are as in the text.
Sample
size is 8-19
fish. (b) Average
EOG
responses
(t
SEM)
of male
goldfish
to water
containing
salineand PGF20-injected
female
goldfish
(wide bars) that has been diluted
10 (0.1;
n = 17) and 100 times
(0.01;
n = 7). Responses
to salineand PGF20-injected
fish water
are compared
by paired
t-tests
(**p0.01).
Average
response
magnitude
(± SEM) is represented
as a percentage
of that induced
by 10’M
L-serine.
The narrow
bars represent
the hypothetical
EOG
responses
that
PGF20-injected
fish water
would
have evoked
if half the injected
PGF20
had been
released
as either
PGF20
(A) or its metabolite
15-keto-prostaglandin
F20 (B; see Fig.
Ic and text).
Their
bases correspond
to the level of responses
evoked
to control
(saline-injected)
female
water.
(c) Semi-logarithmic
plot of the EOG responses
of male goldfish
to PGF20,
its metabolites
and its
precursor,
arachidonic
acid.
Average
response
magnitude
(± SEM) is represented
as a percentage
of that induced
by 10’M
L-serine.
These
data were
collected
from
the same
fish used in Figure
la and
Ib, and the values
plotted
for PGF20
are the same.
Sample
size is 8-19.
Pheromone
and
Release
Olfactory
by PGF2a-Injected
Sensitivity
to PGF2a
PGF2ainjected
fish
greater
LOG
responses
injected
fish (p0.01
for
Goldfish
could
not explain
the magnitude
of
If half of the PGF2a
had been released
Metabolites
water
consistently
evoked
than
water
from
salineboth 0.1 and 0.01 dilutions),
confirming
that PGF2a
injection
a potent
olfactory
stimulant(s)
simple
release
and/or
leakage
evokes
the release
of
(Fig.
ib).
However,
of the injected
PGF2a
water
(a liberal
estimate;
see Table
1) it would
had a concentration
of 10-8 M. Accordingly,
the
dilution
of PGF2a-injected
fish
water
would
contained
should
elicited
have
by
109M
PGF2a,
and
EOG
responses
been
only
marginally
larger
than
saline-injected
fish.
Similarly,
the
concentration
have
contained
able concentration
lmV
_L1_L1_
Ii
III
-8
-8
PG F20
male
sure
and
ized.
each
division
=
5s.
released
other
recorded
by
than
to
found
to
an
15K-PGF2a
of a mature
longer
expocomponent)
to be visualTime
signals,
to it
those
0.01
be
PGF2a-injected
fish
PGF2a.
PGF2a
metabolites
exceedingly
potent
of 1012
M, 100
ic and 3). EOG
to
approach
at
a concentration
of
i0
M, where
responses
slightly
larger
than
10
M Lshape of the LOG response
to 15K-PGF2
rounded
(reflecting
a slower
response)
than that elicited
of the
PGF2a
metabolized
ences
between
have
0.10
have
PGF2a-injected
fish
water
would
only
1010M
PGF2a,
an undetect(Fig. ib). These
findings
suggested
that
the pheromone
contained
components
EOG
responeses
saturation
they
evoked
serine.
The
was
more
1 5-Keto-PGF20
FIG. 2. Electro-olfactogram
recording
(EOG)
responses
goldfish
to 108M
PGF20
and 105M
1SK-PGF20.
A
time
(30 a) was used
to allow
the phasic
(initial
sharp
tonic
(subsequent
plateau)
components
of the EOGs
Responses
were
recorded
on curvilinear
graph
paper.
of
odorant
with
a detection
threshold
times
lower
than that of PGF2a
(Figs.
responses
to
15K-PGF2a
appeared
IiIIIIIIIIIJ._1__1_i._1_1_
I I
EOG
responses.
directly
to the
by PGF2a
(Fig. 2). In theory,
if half
injected
into
female
fish had
been
and released
the EOG
as 15K-PGF2a,
responses
elicited
the
by
differPGF2a-
PROSTAGLANDIN
PHEROMONE
IN
GOLDFISH
(a)
-__
-5
L-SER
CON
-8
17,20P
-10
CON
1045
15K.PGF2
-7M
Response
_J__
-9
-8
0
-7t4
-7
PROSTAGLANDIN
(mV)
1
.:
(b) -8M
.-‘
PGF20c
Response
2
3
#{149}
‘
0
-SM
PGF2cz
(mV)
1
j
2
3
15K
]
DPro
F20
OPre
Duflg
-
-SM
E
-11
-9
-10
-8
Cl)
-7
P.&O1
-SM
TCA
.#{231}__
-8M
4-
1 5-KETO-PROSTAGLANDIN
F20
L-Sar
-7M
17,20P
-SM
#{149}
P=o.o5
P.*O.O1
L-Ser
-7M
-12
CON
P.*o.05
*
**
TCA
17.20P
0)
8)
I-
-7M PGE2
#{149}
-7MPGFI
#{149}
-7M PGF3ca
CON
13,1
FIG.
sponses
taglandin
17,20P,
are not
methanol
-8
-9
1 5-KETO-PROSTAGLANDIN
4-DIHYDRO-
5s.
saline-injected
fish
ib).
13,14-15K-PGF2a
olfactory
stimulant
PGF2a
and similar
acids;
it could
imately
with
to that
not
by PGF2a-injected
weak
olfactory
water
the
fish.
stimulant,
PGFaa
-7M
PGF2
less
and
EOG
than
other
injected
fish
15K-PGF20.
t-tests
(pO.05;
response
Arachidonic
acid
with a threshold
of
In
the
Receptors
Water and
cross-adaptation
elicited
that
was a very
of approx-
both
and
**p0.01).
“Pre’
15K
-SM
15K
#{149}
#{149}
L-serine;
TCA
=
taurocholic
acid;
15K
responses
were
compared
by paired
Sample
size ranged
from
4 to#{243}fish.
=
=
“During’
15K-PGF2a
table
to olfactory
PGF2a
and
tO
15K-PGF2a
PGF2a-injected
PGF2a
M PGF2a
elicited
(Fig.
are not
by these
receptors
4; the
shown),
compounds
that
data
for
suggesting
are attribu-
are most
sensitive
to
15K-PGF2a.
Lastly,
although
adaptation
and PGF2a
reduced
responses
elicited
exposure
to
adaptation
lower
concentrations
never
concentrations
experiments,
and
to 10
responses
PGF,
PGFs
eliminated
of the
other
of
responses
the
other
to equal
PGF.
neither
107M
15K-PGF2a
nor adaptation
to
decreased
the
EOG
responses
to
fish water
(Fig. 4). In contrast,
adaptaand 15KPGF2a
reduced
(p0.05)
the
by
L-Ser
water;
adaptation
by
adaptation
to
108M
PGF2a
saline-injected
tion to PGF2a
#{149}
-9M
FIG.
4. Electro-olfactogram
recording
(EOG)
responses
elicited
(in my)
prior
to adaptation
(light-colored
bars) and later during
adaptation (shaded
bars)
to either
(a) 10’M
15K-PGF20,or(b)
10’MPGF20.
Stimuli
concentrations
are log molar.
Abbreviations
are as given
in the
text
with
the following
exceptions:
Sal-inj
=
0.10
solution
(10%
dilution)
of saline-injected
fish water;
PG-inj
=
0.10
solution
of PGF20-
that of
amino
107M.
Specificity
of Olfactory
to PGF2a-Injected
Fish
15K
-iON
would
be explained
was a relatively
weak
a potency
of L-serine
explain
PGF2
-SM
F20
Representative
electro-olfactogram
recording
(EOG)
reof a mature
male
goldfish
to log molar
concentrations
of pros(PG)
F20,
15K-PGF20,
13, 14-15K-PGF20,
L-serine
(L-Ser),
and controls
(Con;
no odor
added).
Responses
to 106M
PGFs
shown
because
a portion
of these
responses
are attributable
to
carrier.
Curvilinear
graph
paper
was used.
Time
signals,
each
=
-SM
-7
3.
division
and
(Fig.
10
responses
evoked
Adaptation
sponses
to
1 5K-PGF2a
to 109M
15K-PGF2a
also reduced
rePGF2a-injected
fish water,
suggesting
that
was an important
constituent
of PGF2a-
fish
water.
Behavioral
Responses
Waterborne
Exposure
evoked
to
PGFs
to
108M
immediate
and
PGF2a
dramatic
and
108M
increases
15K-PGF2a
in swimming
and social
activity
(nudging)
of male goldfish,
which
coincided
with decreases
in feeding
(Fig. 5). Exposure
to ethanol
control
did not affect
the behavior
of male
goldfish.
The
activity
evoked
by waterborne
PGFs
injected
fish water
(data
not shown).
In other
cross-adaptation
experiments,
responses
to
10M
L-serine,
107M
taurocholic
acid, and 108M
17,20P
were
not
reduced
by adaptation
to either
was characterized
by chasing,
a behavior
characteristic of sexual
arousal
(data
not shown).
The behavioral
thresholds
for responses
elicited
by PGF2a
was 10-8 M
and
10#{176}M for 15K-PGF2a.
There
was no apparent
PGF2a
responses
exposure
abolished
or
15K-PGF2a
to PGF1a,
during
(Fig.
PGF3a,
adaptation
4).
Conversely,
and
PGE2
to
equal
were
concentrations
EOG
all
difference
both
in the
to
compounds
nature
PGF2a
of the
and
were
responses
15K-PGF2a.
relatively
elicited
Responses
short
in duration,
by
to
SORENSEN
1046
PGF2
U
20
30
500
>
=
400
Pre
300
During
E
*
E
ioo
Ci)
0
**
*
much
higher
serum
PGF2a
employed
metabolites.
PcO.01
Molar
i:J[t
latory
Concentration
30
*
may
i:J
Molar
ijj
pound),
of
4,
4,
xc
_
30-I
I
4,1
20
I
i0.I
I
C
E
rr
100
___9
Cl)
Log
FIG.
centrations
z10J
.
Molar
5. Behavioral
responses
of male
of waterborne
prostaglandins
0Ir72-8
within
30 mm
10
(data
mm
not
U.
[1
centrally
results).
conan
olfactory
odorants;
suggest
01
.8
Concentration
fish exposed
to different
(PGF20,
15K-PGF20)
ethanol
control
(ETOH).
Median
response
values
are
centrations
are log molar.
Because
of limited
space,
feeding
is not shown
for concentrations
less than
same
trends
were
apparent.
Results
were
analyzed
test
for
dependent
samples
(paO.05;
‘p001;
5 fish for each concentration).
peaking
within
I
and declining
shown).
and
shown.
Stimuli
condata on nudging
and
10
M although
the
by the signed
ranks
n = 11 groups
of
to
basal
levels
DISCUSSION
This
study
establishes
F prostaglandins
component
of the goldfish
postovulatory
To our
knowledge,
this is the first
release
PGFs
to their
environment.
goldfish
release
considerable
quantities
active
PGF
to
the
water
and
cease
as a critical
pheromone.
report
that
fish
Ovulated
female
of immunorereleasing
these
compound(s)
shortly
after
spawning
when
they
are
no longer
reproductively
active.
Furthermore,
PGF2injected
female
goldfish,
which
are known
to release
odorant(s)
with
actions
similar
to those
of the
postovulatory
pheromone
also release
immunoreactive
actual
quantity
of PGFs
(Sorensen
et al., 1986),
PGF.
It is likely
that the
released
by ovulated
fish is
1987).
tracts
as
the
anti-
Although
washings
of
Sorensen
(see
Stacey
P.
et a!.,
contributing
known
olfactory
appear
to
and
their
acids
first
the
LOG
W.,
1986),
source.
bile
somewhat
low
do
were
detection
only
goldfish
acids
prostaglandins
less
to PGFs
size,
of
stimulants
in a vertebrate.
possess
at least two classes
other
to staturate;
odorant
in
and
report
PGFs,
one highly
specific
(or a similar
unknown
responses
large
tendency
potent
Controls
400
developed.
in crude
1979;
fish
receptors
for
to 15K-PGF2a
PGF2a.
30
because
and several
studies
have suggested
is an important
source
of postovuin
is the
olfactory
sensitive
Concentration
500
being
is found
be an important
functioning
Male goldfish
**
Log
here
20
This
**
report
very
low cross-reactivity
with
An
antiserum
for
15K-PGF2a
(Bouffard,
results),
fluid
pheromones
urine
**
we
had
currently
PGF
eggs
unpublished
that ovulatory
F2
.500
EtoH
is
ovulated
Log
than
compound
immunoreactive
P’cO.05
.r.r!
1 5K-PG
:20
**
ET AL.
specific
notable
17, 20P represents
and responses
not
saturate
transduced
(Sorensen
Neural
responses
to
to
because
thresholds,
and
to
a more
amino
(Sorensen
Electrical
recording
from
the
has
confirmed
that
responses
of
and
com-
medial
to
et al.,
olfactory
PGFs
are
P. W., unpublished
PGFs
are similar
to
responses
transduced
in response
to other
there
is currently
no empirical
evidence
to
that
a specialized
neural
system
such
as the
terminal
nerve
ates responses
Because
both
(Demski
and Northcutt,
1982)
to the postovulatory
pheromone.
classes
function
independently
other
known
categories
of
PGF
of receptors
of odorants,
olfactory
medi-
receptors
that respond
prostaglandins
represent
a fourth,
previously
unsuspected,
of potent
olfactory
stimulants
for
fish.
both
cross-adaptation
(this
study)
and
experiments
(Sorensen
P. W., unpublished
to
category
Although
mixture
results)
suggest
that
PGF
olfactory
receptors
specifically
recognize
PGF2a
and
15K-PGF2a
at low concentrations,
considerable
cross-reactivity
was evident
when
concentrations
of 10
M and greater
were tested.
We
believe
that the dramatic
increase
in the LOG evoked
by
106M
15K-PGF2a
in the
water
was
caused
by
cross-reacting
PGF2a
receptors.
However,
because
levels
of PGF2
in the water
and
serum
of
ovulated
goldfish
(this study;
Bouffard,
1979)
are less
than
10
M, olfactory
to distinguish
naturally
PGFs.
This
study
goldfish
release
specificity
occurring
confirms
that
potent
odorant(s)
should
be adequate
concentrations
of
PGF2a-injected
which
female
are detected
PROSTAGLANDIN
by
those
olfactory
receptors
that
Although
the size of the
water
from
PGF2a-injected
that
a PGF2a
metabolite
important
component
mone’s
exact
chemical
to reduce
(Van
17,20P
is
pheromone.
not
washings
et
PGF2
sponses.
odorants
et
a!.,
been
1980;
1985;
Bryant
1988)
1984;
Hara
et a!.,
and
Atema,
of
exhibited
exposed
The
(108M
Saglio
nor
to
thresholds
for PGF2a
a!.,
1980)
that
re-
these
(Doving
Fauconneau,
although
it is
PGFs
similar
to
fish, and that the
(Sorensen
et a!.,
of similarity
analysis.
has
yet
to
waterborne
observed
of ovulated
et
and
fish release
ovulated
are similar
and
in
males
PGF2a-injected
1976;
for
and
behavioral
responses
10’#{176}M for 15K-PGF2a)
chasing
(Partridge
it is likely
that
be
PGFs
al.,
were
approximately
an order
than
the thresholds
determined
probably
due to odor
dilution
effectiveness
of our
experimental
that repeated
brief exposure
to
stimulate
strong
behavioral
goldfish
spawning
behavior
extensive
observations),
that
control
Lastly,
behaviors
odors
(Partridge
et
these
suggested
mixtures
1987).
exposed
same
to the
goldfish
1986).
PGFs
males
the
suggest
Doving
1984;
exact
degree
by chemical
Groups
release
the postovulatory
to 1 5K-PGF2
for
authors
have
to pheromonal
clear that PGF2-injected
those
released
by naturally
actions
of these
odorants
1986),
the
determined
of
to PGFs
fact that
17,20P
strongly
responsible
Various
contribute
et a!.,
a!.,
stimulate
by
decreased
the EOG
responses
fish
water,
suggesting
that
the
bile acids commonly
found
in fish
and
(Hara
have
not
is an
phero-
verification
failure
of adaptation
to 17, 20P and the
a component
Neither
adaptation
adaptation
to
to
saline-injected
L-amino
acids
could
requires
does
Kraak
PGFs.
elicited
by
indicated
(probably
15K-PGF2)
of the pheromone,
the
analysis.
The
EOG responses
Der
to
LOG
responses
fish
strongly
composition
PGF2a-injection
respond
PHEROMONE
Sorensen
et
a!.,
to
of
magnitude
higher
by EOG
recording,
during
injection.
The
design
indicates
wisps
of an odor can
responses.
Because
is characterized
by
et a!., 1976;
males naturally
personal
encoun-
ter traces
of female
odor
plumes
in a manner
similar
to that tested
here.
The short
duration
of the behavioral
response
to waterborne
PGFs
could
indicate
that,
as suggested
by Partidge
et a!. (1976),
the social
context
of pheromone
exposure
is important;
behavioral
feedback
female
may
Goldfish
may
be
(visual,
tactile,
essential
for
have
evolved
etc.)
from
a prolonged
to
ignore
a receptive
response.
the
presence
IN
GOLDFISH
of lingering
Unfortunately,
to address
1047
pheromones
our simple
the
specific
relevance
of waterborne
PGFs
may have different
ity
that
goldfish
gender-specific
tests
of
future.
It is
goldfish
mones
after
spawning
has ceased.
bioassay
does not allow us
behavioral/endocrinological
PGFs,
the
functions,
that the
possibil-
also release
species-specific
and/or
chemical
cues.
More
sophisticated
pheromone
function
now
apparent
sequentially
that
release
that
possibility
and the
have
are
planned
for
periovulatory
two
hormonal
important
and
the reproductive
physiology
Through
their
initial
actions
quent
roles
as pheromones,
differing
the
female
pheroeffects
and behavior
of
as hormones
and
these
compounds
on
males.
subsesyn-
chronize
female
and
male
reproductive
physiology
and
spawning
behavior.
Accordingly,
the
reproductive
physiology
and
behavior
of spawning
male
goldfish
can
be modeled
as a “dual
pheromone
system”
ogy of ovulatory
Many
about
both
based
on
females
important
questions
this dual
pheromones
analysis
and
of their
release
particularly
metabolism
principle
PGF2.
remain
pheromone
require
both
the
have
temporal
yet
to
the
postovulatory
Because
endocrinolto
system.
The
verification
be answered
identities
of
by chemical
pattern
and
be determined.
important
to
in goldfish
to
metabolite/pheromone
Similarly,
component
determined.
the reproductive
(Fig. 6).
examine
determine
is
method
It will
be
prostaglandin
whether
the
actually
15K-
significance
of
pheromone
also
15K-PGF2a
saturates
a
multi-
has to be
at a low
concentration,
its ability
to function
as a close-range
signal
is probably
limited;
it may act as a long-range
cue signifying
the presence
of an ovulated
female(s),
while
fying
PGF2a
functions
the
ovulated
that these
pheromones
Finally,
because
males
females
experience
gonadotropic
hormone
is possible
that
one
as a close-range
individual.
It is
synergize
spawning
each
with
signal
identialso
possible
other’s
actions.
PGF2a-injected
rapid
elevations
in circulating
and milt (Kyle
et a!., 1985),
of these
PGFs
has an endocrine
it
function.
Our
discovery
that
both
the
preovulatory
and
postovulatory
pheromones
in goldfish
are hormones
with
reproductive
functions
in a variety
of fish
(Goetz,
Canario,
theoretical
1983;
1987;
Stacey
Goetz
arguments
and Goetz,
et a!., 1987)
that
sex
1982;
strongly
hormones
Scott
and
reinforces
and
their
1048
SORENSEN
ET
AL.
highly
specific
Sorensen
glandins
GD
hormonal
FEMALES
or
achieve
(RELEASE
TO
THE
appear
+
1200
=
2000
0400
Time
1200
of Day
FIG.
6. A model
of the dual sex pheromone
system
employed
by
goldfish.
Environmental
cues
trigger
an ovulatory
surge
in gonadotropic hormone
(GtH)
in vitellogenic
females
in the afternoon
in the late
spring
(Stacey
et al.,
1979),
which
subsequently
stimulates
17,20P
synthesis
by the
ovary
(Kobayashi
et al.,
1987).
Hormonal
17,20P
induces
final
oocyte
maturation
(resumption
of meiosis;
Nagahama
et
al., 1983)
and is released
to the water
where
it functions
as a preovulatory
priming”
pheromone.
This pheromone
evokes
a surge
in circulating GtH
in males
(Dulka
et al., 1987),
which
stimulates
the synthesis
of testicular
17,20P.
Elevated
hormonal
17,20P
in turn
evokes
an increase
in milt
production
by
the
time
of ovulation
and
spawning
(Dulka
et al., 1987).
Later
at the time of ovulation,
females
produce
F
type
prostaglandins
(PGFs)
to mediate
follicular
rupture
and to trigger
female
spawning
behavior
(Stacey
and Goetz,
1982).
Circulating
PGFs
are subsequently
metabolized
and
released
to the water
where
they
function
as a postovulatory
pheromone
that
stimulates
male
sexual
arousal,
effecting
spawning
synchrony.
their
metabolites
are
pheromones.
species-specificity.
This
commonly
used
as
raises
important
We believe
that
reproductive
questions
because
about
phero-
mones
are greatly
diluted
under
natural
conditions
selective
pressure
for the evolution
of species-specific
pheromones
probably
only exists
among
species
that
spawn
in
close
proximity.
Both
species-specific
(Liley,
1982;
Honda
1982a,
b; McKinnon
and Liley,
1987)
and
nonspecies-specific
1965;
Chen
and
Martinich,
McKinnon
and Liley,
1987)
in fish,
Because
although
goldfish
(Hunter
1975;
pheromones
and Hasler,
Rossi,
1979;
have
been
their
identities
are
olfactory
receptors
be
in
common
1982).
perceived
cues
unare
are
selected
prostaglandin
among
Finally,
within
(chemical,
metabolites
effectively
mammals
pheromona!
the context
auditory,
also
that
used
visual,
provide
essential
prostaglandins
as
postovulatory
by many
externally
fertilizing
teleosts,
spawning
in these
species
coincides
with
and prostaglandins
appear
to play
a funrole in modulating
ovulation
(Dennefors
et
a!., 1983;
Goetz,
1983;
Stacey
et a!., 1987).
have
already
been
shown
to trigger
female
behavior
in a variety
of externally
fertilizing
(Stacey
Liley
and
and
Stacey,
also
and
PGF2a
pheromone
promelas;
believe
males
and
preovulatory
soon
after
hormone
pheromone.
PGFs
sexual
species
Goetz,
1982;
Cole
and Stacey,
Tan,
1985;
Villars
and
Burdick,
1987),
to elicit
(Pimephales
that
injection
release
Cole
are
Thus,
fertilizing
closely
and whose
ovulation,
are
or metabolite
been
found
in the fathead
minnow
and
Smith,
1987).
We
externally
females
period,
has
1984;
1986;
fish
associated
females
whose
during
the
must
spawn
likely
to use a maturational
as a preovulatory
priming
although
female
sex
pheromone
systems
in fish may
differ
in detail,
it is likely
that
most
species
use
similar
hormonal
compounds.
It
seems
ironic
that
the original
definition
of a pheromone
reported
known.
behavioral
and/or
Male
Sex)
Behavior
+
(Milt)
to
1987;
prostain either
where
differences
and Kindahl,
is probably
al.,
and
could
species-specificity
(Granstrom
information
other
et
results)
variations
pathways
Species-specific
pheromones
because
ovulation,
damental
e-17,20P-b.
(Sorensen
electrical,
and tactile)
that may
species-specific
information.
We believe
it is highly
probable
Spawning
Synchrony
MALES
metabolic
metabolism
of
WATER)
steroids
pheromonal
for.
ixo
for
P. W.,
unpublished
(this
study),
minor
stated
that
“unlike
hormones
...
the
substance
is not secreted
into the blood
but outside
the body;
it
does not serve humoral
correlation
within
the organism but communication
between
individuals”
(Karlson and Luscher,
1959).
It now appears
that, at least
in the goldfish,
hormones
and reproductive
pheromones
can be one and the same.
ACKNOWLEDGMENTS
We thank
K. J. Chamberlain
iments
and G. 0. Sorensen
for
Peter
for his enthusiastic
support
for her help with
her editorial
advice.
of these
studies.
the behavioral
We also thank
experR. E.
REFERENCES
Bouffard
tion,
RE,
1979.
ovulation
The
role
of
and spermiation
prostaglandins
in the
during
goldfish,
sexual
Carassius
maturaauratus.
PROSTAGLANDIN
PHEROMONE
M. Sc, Thesis,
Univ.
British
Columbia,
vancouver,
BC, p. 155
CM Jr, Rosen
DE, 1966.
Modes
of Reproduction
in Fishes.
New
York:
Natural
History
Press,
pp. 221-24
Bryant
BP, Atema
J, 1987. Diet manipulation
affects
social
behavior
of
catfish:
the importance
of food
odor.
J Chem Ecol 13:1645-62
Caprio
J, 1984, Olfaction
and taste
in fish.
In: Bolis
L, Keynes
RD.
Maddrell
SMP (eds.),
Comparative
Physiology
of Sensory
Systems.
New York:
Cambridge
University
Press,
pp. 257-83
Caprio
J, Bryd RP Jr, 1984. Electrophysiological
evidence
for acidic,
basic,
and neutral
amino
acid olfactory
receptor
sites in catfish.
J
Gen Physiol
84:403-22
Cetta
F, Goetz
FW,
1982.
Ovarian
and plasma
prostaglandin
E and F
levels
in brook
trout
(Salvelinius
fontinalis)
during
pituitary
Breder
Chen
Cole
induced
ovulation.
Biol Reprod
27:1216-21
LC, Martinish
RL,
1975.
Pheromonal
stimulation
inhibition
of ovulation
in the zebrafish,
Brachydanio
Bull 73:889-94
KS,
Smith
RJF,
1987.
Release
treated
female
fathead
minnows,
lates male
courtship.
Horm
Behav
of
Pimep
chemicals
by
and
metabolite
rerio.
Fish
prostaglandin
that stimu-
hales prom elas,
21:440-56
Cole
KS, Stacey
NE,
1984.
Prostaglandin
induction
of spawning
behavior in Cichiasoma
bimaculatum
(Pisces
Cichlidae).
Horm
Behav
18:
23 5-48
Colombo
L, Belvedere
PC, Marconato
A, Bentivenga
F, 1982.
Pheromones
in teleost
fish. In: Richter
CJJ, Goos
TJT (eds.),
Proceedings
of the
Second
International
Symposium
on the
Reproductive
Physiology
of Fish. The Netherlands:
Pudoc,
pp. 84-94
Demski
LS, Northcutt
RG,
1983.
The
terminal
nerve:
a new chemosensory
system
in vertebrates?
Science
202:435-37
Dennefors
B, 1-lanberger
L, Hillensjo
T, Holmes
P, Janson
P0, Magnusson
C, Nilsson
L, 1983.
Aspects
concerning
the role
of prostaglandins
for ovarian
function.
Acta
Obstet
Gynecol
Scand
Suppl
113 :31-41
Doving
KB,
1976.
Evolutionary
trends
in olfaction,
In: Benz
G (ed),
Structure-Activity
Relationships
in Chemoreception.
London:
IRL
Press,
pp. 149-59
Doving
KB, Selset
R, Thommesen
G, 1980.
Olfactory
sensitivity
to bile
acids
in salmonid
fishes.
Acts
Physiol
Scand
108:123-3
1
Dulka
JG,
Stacey
NE,
Sorensen
PW, Van
Ocr Kraak
GJ,
1987,
Sex
steroid
pheromone
synchronizes
male-female
spawning
readiness
in the goldfish.
Nature
(Lond)
325:251-53
Evans
RE, Hara TJ, 1985.
the characteristics
of the electro-olfactorgram
(EOG):
its loss and recovery
following
olfactory
nerve
section
in
rainbow
trout
(Salmo gairdneri).
Brain
Res 3 30:65-75
Getchell
1W,
1974.
Electrogenic
sources
of slow
voltage
transients
recorded
from
frog
olfactory
epithelium.
J Neurophysiol
37:
1115-30
Goetz
FW,
1983.
Hormonal
control
of oocyte
final
maturation
and
ovulation
in fishes.
In:
Hoar
WS,
Randall
DJ,
Donaldson
EM
(eds.),
Fish Physiology,
Vol IXB.
New
York:
Academic
Press,
pp.
117-70
Goetz
FW
Ranjan
M, Berndtson
AK, Duman
P, 1987.
The mechanism
and
hormonal
regulation
of ovulation:
the role of prostaglandins
in teleost
ovulation.
In: Idler
DR,
Crim
LW, Walsh
JM,
(eds.),
Proceedings
of the Third
International
Symposium
on the Reproductive
Physiology
of Fish.
St. Johns,
Newfoundland:
Memorial
University
Press,
pp. 235-38
Granstrom
E,
Kindahl
I-I, 1982.
Species
differences
in circulating
prostaglandin
metabolites:
relevance
for the assay
of prostaglandin
release.
Biochim
Biophys
Acts
713:555-69
I-tars
Ti,
1982.
Structure-activity
relationships
of amino
acids
as
olfactory
stimuli.
In:
Hara
Ti (ed.),
Chemoreception
in Fishes,
New York:
Elsevier,
pp. 135-58
Hara
Ti,
1986.
Role of olfaction
in fish behaviour.
In: Pitcher
TJ (ed),
Behaviour
of Teleost
Fishes.
London:
Croom
Helm,
pp. 152-76
Hara
TJ, Macdonald
S. Evans
RE, Marui
T, Arai S, 1984.
Morpholine,
bile acids
and skin
mucus
as possible
chemical
dues
in salmonid
homing:
electrophysiological
re-evaluation,
In:
McCleave
JD,
Arnold
GP, Dodson
JJ, Neill WH (eds.), Mechanisms
of Migration
in Fishes.
New York:
Plenum
Press,
pp. 363-78
Honda
H, 1982a.
On the female
sex pheromones
and courtship
behavior in the bitterling
Rhodeus
ocellatus
ocellatus
and Acheilogna-
thus
lanceolatus.
Bull
Jpn
Soc
Sci
Fish
48:43-45
IN
1049
GOLDFISH
Honda
H, 1982b.
On the female
sex pheromones
and courtship
behavior in the salmonids
Oncorhynchus
mason
and 0. rhodurus.
Bull
Jpn Soc Sci Fish 48 :47-58
Hoult
JRS,
Moore
PK, 1977.Pathways
of prostaglandin
F2a metabolism
in mammalian
kidneys.
BrJ Pharmacol
61 :615-26
Hunter
JR,
shiner,
Hasler
Notropis
cyanelleus.
AD,
Copeia
P, Luscher
M,
biologically
active
Kittredge
JS, Terry
M,
the molting
hormone
Karlson
crassipes,
Cancer
1965.
umbratilis,
Spawning
and
the
association
green
of
sunfish,
the
redfish
Lepomis
1965:265-81
1959.
“Pheromones”:
a new term
for a class of
substances.
Nature
(Lond)
183:55-56
Takahashi
R, 1971,
Sex pheromone
activity
of
crustecdysone
on male
crabs
(Pachygrapsus
antennarius
and
C. anthony).
Fish
Bull
69:
337-43
Kobayashi
M, Aida
K, Hanyu
I, 1987.
Hormone
changes
during
ovulation
and
effects
of steroid
hormones
on plasma
gonadotropin
levels
and ovulation
in goldfish.
Gen Comp
Endocrinol
67:24-3
2
Kyle
AL, Stacey
NE, Peter
RE, Billard
R, 1985.
Elevations
in gonadotropin
concentrations
and milt
volumes
as result
of spawning
in
the goldfish.
Gen Comp
Endocrinol
57:10-22
Lambert
JGD,
van den Hurk
R, Schoonen
WGEJ,
Resink
JW, van Oordt
PGWJ,
1986.
Gonadal
steroidogenesis
and
the
possible
role
of
steroid
glucuronides
as sex pheromones
in two species
of teleosts.
Fish Physiol
Biochem
2:101-07
Lewis
DH, Tarpley
RJ, Marks
JE, Sis RF, 1985.
Drug induced
changes
in the
olfactory
organ
of channel
catfish
ictalurus
punctatus
Rafinesque.
J Fish Biol 26:355-58
Liley NR,
1982.
Chemical
communication
in fish. Can J Fish Aquat
Sci
39:22-35
Liley
NR,
Tan
ESP,
1985.
The
induction
of spawning
behaviour
in
Punctius
gonionotus
(Blecker)
by treatment
with
prostaglandin
PGF20.
J Fish Biol 26:491-502
McKinnon
JS,
Liley
NR,
1987.
Assymetric
species
specificity
in response
gaster
to female
pheromone
(Pisces:
Blontidae).
Can
by
J
males
of two
zool
65:1129-34
species
of
tricho-
Nagahama
Y, Hirose
K, Young
G, Adachi
5, Suzuki
K, Tamaoki
B,
1983.
Relative
in vitro
effectiveness
of 17o,
20j3-dihydroxy-4pregnen-3-one
and other
pregnene
derivatives
on germinal
vesicle
breakdown
in oocytes
of
ayu
(Plecoglossus
altivelis),
amago
salmon
(Oncorhynchus
rhodurus),
rainbow
trout
(Salmo gairdneri),
and
goldfish
(Carassius
auratus).
Gen Comp
Endocrinol
51:15-23
0,
1971.
The
electro-olfactogram.
In:
Beidler
LM
(ed),
Chemical
Sources,
1. Olfaction,
Handbook
of Sensory
Physiology,
Vol IV/1.
New York:
Springer-Verlag,
pp. 95-131
Partridge
BL, Liley
NR, Stacey
NE, 1976.
The role of pheromones
in
the sexual
behavior
of the goldfish.
Anim
Behav
24:291-99
Rossi
AC,
1969.
Chemical
signals
and
nest
building
in two species
of
Colisa
(Pisces:
Anabantidae).
Monit
Zoo! Ital 3:225-37
Saglio
P Fauconneau
B, 1985.
The
amino
acid
content
in the skin
mucus
of goldfish,
Carassius
auratus:
influences
of feeding.
Comp
Biochem
Physiol
82A:67-70
Samuelsson
B, Granstrom
E, Green
K, Hamberg
M, Hammarstrom
5,
1975.
Prostaglandins.
Annu
Rev Biochem
44:669-95
Scott
AP, Canario
AVM,
1987.
Status
of oocyte
maturation-inducing
steroids
in teleosts.
In:
Idler
OR,
Crim
LW, Walsh
JM, (eds.),
Proceedings
of the Third
International
Symposium
on the Reproductive
Physiology
of Fish.
St. Johns,
Newfoundland:
Memorial
University
Press,
pp. 224-34
Sorensen
PW, Hara
TJ, Stacey
NE, 1987.
Extreme
olfactory
sensitivity
of mature
and gonadally-regressed
goldfish
to a potent
steroidal
pheromone,
1 7,
20$3-dihydroxy-4-pregnen-3-one.
J Comp Physiol
A 16:305-13
Sorensen
PW, Stacey
NE,
Naidu
F, 1986.
Release
of spawning
pheromone(s)
by naturally
ovulated
and prostaglandin
injected,
nonovulated
female
goldfish.
In: Duvall
D, Muller-Schwarze
0, Silverstein
RM
(eds.),
Chemical
Signals
In Vertebrates,
Vol
IV.
Ecology,
Evolution
and Comparative
Biology.
New York:
Plenum
Press,
pp.
149-54
Stacey
NE,
1976.
Effects
of indomethacin
and prostaglandins
on the
spawning
behavior
of female
goldfish.
Prostaglandins
12:113-26
Stacey
NE, 1981.
Hormonal
regulation
of female
reproductive
behavior
in fish. Am Zool
21:305-16
Stacey
NE, 1987.
Roles
of hormones
and pheromones
in fish reproduc-
Ottoson
1050
SORENSEN
tive
Stacey
Stacey
Stacey
Stacey
Stacey
Stacey
Stacey
Stacey
behavior.
In:
Crews
D (ed),
Psychobiology
of
ET
AL.
Reproductive
Behavior.
New York:
Prentice-Hall,
pp. 28-69
NE, Cook
A, Peter
RE, 1979.
Ovulatory
surge of gonadotropin
in the goldfish.
Gen Comp
Endocrinol
37:246-49
NE, Goetz
FW, 1982.
Role
of prostaglandins
in fish reproduction.
Can J Fish Aquat
Sci 39:92-98
NE, Kyle
AL, 1983.
Effects
of olfactory
tract
lesions
on sexual
and feeding
behavior
of fish. Physiol
Behav
30 :621-28
NE, Kyle
AL, Lucy
NR, 1986.
Fish reproductive
pheromones.
In: Duvall
D, Muller-Schwarze
D, Silverstein
RM (eds.),
Chemical
Signals
in Vertebrates,
Vol IV. Ecology,
Evolution
and Comparative Biology.
New York:
Plenum
Press,
pp. 119-3
3
NE,
Liley
NR,
1974.
Regulation
of spawning
behaviour
in the
female
goldfish.
Nature
(Lond)
247:71-72
NE, Peter
RE, 1979.
Central
action
of prostaglandins
in spawning
behavior
of female
goldfish.
Physiol
Behav
22:1191-96
NE, Sorensen
PW, 1986.
17o, 203-Dihydroxy-4-pregnen-3-one:
a
steroidal
primer
pheromone
which
increases
milt
volume
in the
goldfish.
Can J Zool 64:2412-17
NE, Sorensen
PW, Dulka
JG, Van Der Kraak
GJ, Hara Ti, 1987.
Teleost
sex pheromones:
In: Idler
DR, Crim
LW,
International
van
van
Van
Villars
recent
Walsh
Symposium
studies
JM (eds.),
on
on identity
Proceedings
the
of Fish.
St.
Johns,
Newfoundland:
pp. 150-54
den
Hurk
R, Lambert
JGD,
1983.
function
as sex pheromones
for male
and function.
of the Third
Reproductive
Memorial
Ovarian
zebrafish,
Physiology
University
steroid
Press,
glucuronides
Bracbydanio
rerio.
Can J Zoo! 61:2381-87
den
Hurk
R, Schoonen
WGEJ,
van
Zoelen
GA,
Lambert
JGD,
1987.
Biosynthesis
of steroid
glucuronides
in the
testis
of the
zebrafish
Brachydanio
rerio,
and
their
pheromonal
function
as
ovulation
inducers.
Gen Comp
Endocrino!
68:179-88
Der Kraak
G, Sorensen
PW, Stacey
NE, Dulka
JG, 1988.
Periovulatory
female
goldfish
release
three potential
pheromones:
17a,20pdihydroxyprogesterone,
1 7a,20a-dihydroxyprogesterone
glucuronide
and
l7cs-hydroxy
progesterone.
Gen
Comp
Endocrinol
(In
press)
TA, Burdick
of the paradise
treatment.
Gen
M, 1986.
Rapid
fish (Macropodus
Comp
Endocrinol
decline
opercularis)
63:157-61
in the
behavioral
response
to prostaglandin
F2a